* Importing data (remember to change repository)

import delimited "repository\data for table 3.csv"

* Note: a few schools were treated with the low performance treatment though 
* they exceeded the aspiration levels. These schools are captured by the over_asp
* variable and are excluded from the analysis

* Running analyses
eststo: quietly reg search2_sd treat if over_asp==0,vce(cluster school)

eststo: quietly reg semi treat if over_asp==0,vce(cluster school)

eststo: quietly reg changesd treat if over_asp==0,vce(cluster school)

ttest changesd if over_asp == 0, by(treat)

eststo: quietly reg surplussd treat if over_asp==0,vce(cluster school)

esttab